Derivation of 24p3^−/− mice is described elsewhere (32). Mice were housed in a pathogen-free facility and all animal protocols were approved by the Institutional Animal Care and Use Committee of Case Western Reserve University. 24p3^−/− mice were maintained on a C57BL/6J genetic background. All experiments were performed using age- and gender-matched 24p3^+/+ mice.

Mice were administered with 4% (w/v) DSS (average molecular weight 40,000 g/mol; PanReac AppliChem, Maryland Heights, MO) dissolved in sterile, distilled water ad libitum for 2 weeks. Fresh DSS solution was replenished every third day.

Mice were examined daily for weight loss and evidence of colitis by examining stool consistency and the presence of occult blood. The baseline clinical score was determined on day 1 post DSS administration.

Mice were anesthetized with 4% isoflurane prior to endoscopy. Colonoscopy was performed on the 7th day of post DSS treatment and inflammation was evaluated by a previously validated endoscopic scoring system (33), which incorporates 4 different parameters to assess colonic inflammation: perianal findings (diarrhea, bloody feces or rectal prolapse), wall transparency (ability to observe colonic mucosal blood vessels), intestinal bleeding (spontaneous or procedurally induced by endoscope due to mucosal friability), and focal lesions (edema, erosions and ulcers). Colonoscopy was performed as described in Di Martino et al. (34) using a flexible digital ureteroscope (URF-V, Olympus America, Center Valley, PA) with an 8.5 Fr (2.8 mm) tapered-tip design and a motion range of 180° in an up angle and 275° in a down angle. The endoscope system includes a video system center (Olympus America), a xenon light source (Olympus America) and a video recorder (Medi Capture, Plymouth Meeting, PA). Sub scores for each parameter ranging from 0 (normal colonoscopy) to 3 (maximum severity of colonic changes) were used to evaluate colonic inflammation. The sum of these sub scores was used to define colonic health as follows: healthy (0–1), mild colitis (2–4), and moderate colitis (5–7).

Whole blood from naïve as well as DSS administered 24p3^+/+ and 24p3^−/− mice was collected by tail vein puncture and blood cell counts were determined in an automated blood counter (IDEXX, Westbrook, ME).

24p3 levels in naïve as well as DSS administered mice were measured using a commercial ELISA kit from (R& D systems, Minneapolis, MN) as per the manufacturer's instructions.

Serum was collected from blood collected by cardiac puncture at indicated time points. Mouse cytokines and chemokines in sera were measured using cytokine 9-Plex discovery (Eve technologies, Calgary, Alberta, Canada).

Total RNA was isolated from liver samples using the Trizol method (Invitrogen). DNase I (Promega) treated RNA was reverse transcribed using Superscript III RT from Invitrogen as per the manufacturer's recommendations. The resulting cDNAs were used for real time PCR analysis using SYBR Green master mix (Promega, Madison, WI) following the manufacturer's recommendations. The fold-change was calculated using the ΔΔCT method.

Iron levels in sera and tissue samples were measured as described (35).

DSS administered mice were injected daily with 10 μg of anti PDGF-BB antibody (EMD Millipore, Burlington, MA) via intraperitoneally. Survival rate was assessed 2 weeks after completion of the experiment.

Fecal samples from naïve and DSS administered mice were collected and DNA was extracted using a commercial kit from Qiagen as per manufacturer's instructions. DNA concentrations were adjusted, standardized and subjected to PCR amplification using a kit from Illumina using primers specific for 16s rRNA gene. Pooled libraries were then assessed for quality and sent for sequencing at Texas A & M university. Sequence analysis was performed in the laboratory of Dr. Jan Suchodolski at Texas A & M as per established procedures (36).

Data are presented as mean ± SD. Error bars shown in figures represent SD. Statistical analyses were performed by one-way analysis of variance and the Tukey HSD (honestly significant difference) test was performed for multiple comparisons using SAS/STAT software. For a two-way comparison, Students t-test was used with Welch's correction. P < 0.05 were considered statistically significant.

The Case Western Reserve University IACUC committee approved all of the animal studies performed.

Oral administration of DSS is toxic to the colonic epithelium and triggers a local inflammatory reaction by exposing the lamina propria of the colon to resident gut flora (4, 5). These invaders also spread systemically via a hematogenous route thus triggering a wide spread acute phase-like reaction. In accordance with previous findings that 24p3 is an acute phase protein (16), we found that 24p3 levels were increased by >1,000-fold at the RNA level and >200-fold at the protein level, respectively in DSS-treated 24p3^+/+ mice (Figures 1A,B). Such an acute increase in 24p3 levels upon DSS treatment suggests that it may play an important role in the inflammatory response to DSS.

DSS-induced Ulcerative Colitis (UC)-like condition is a widely accepted model to uncover phenotypes in gene knockout mice (37, 38). Age- and sex-matched 24p3^+/+ and 24p3^−/− mice were exposed to 4% DSS in their drinking water to induce a UC-like condition. We found that DSS-administered 24p3^+/+ mice rapidly lost weight and succumbed to colitis with an onset of death as early as 7 days post DSS exposure with nearly all mice dying within 10 days after administration of DSS (Figures 1C,D). In contrast, only 40% of DSS-administered 24p3^−/− mice died on day 10 compared to the 24p3^+/+ cohort suggesting that these mice are relatively resistant to DSS-induced colitis (Figure 1C). Eventually, all DSS-administered 24p3^−/− mice succumbed to DSS-induced colitis at day 15 (Figure 1C). In addition, we also found that DSS treated 24p3^+/+ mice rapidly lost weight whereas the loss of body weight was more gradual in 24p3^−/− mice (Figure 1D). The loss of body weight is concomitant with survival in both groups of DSS treated mice. Differences in rectal bleeding were also apparent between the two groups. Gross examination of colon by endoscopy revealed hemorrhagic diarrhea in DSS treated 24p3^+/+ mice whereas the severity of bleeding was significantly lower in DSS treated 24p3^−/− mice (Figure 1E). A quantitative assessment of colonoscopy findings also yielded significant differences between 24p3^+/+ and 24p3^−/− mice upon DSS exposure (Figure 1F).

These clinical assessments were validated histologically using representative colon sections. As expected, we observed marked histopathological changes in hematoxylin & eosin (H&E)-stained colons from DSS-administered 24p3^+/+ mice characterized by crypt loss, infiltrating leukocytes, ulceration, and submucosal edema (Figure 1G). In contrast, colon sections from DSS-administered 24p3^−/− displayed mild to moderate inflammation (Figure 1G). Consistent with the absence of disease in naïve mice that were not administered DSS, no signs of inflammation or tissue damage were observed in their colons (Figure 1G). Semiquantitative scoring of these histological sections confirmed that colitis severity in DSS-administered 24p3^+/+ mice was higher than DSS-administered 24p3^−/− mice (data not shown). In summary, DSS-administered 24p3^+/+ mice bore all the hallmarks of colitis and succumbed to DSS treatment while 24p3^−/− mice exhibited a reduced sensitivity to DSS-induced colitis. These results demonstrate that 24p3 plays a role in the progression of DSS-induced colitis.

24p3 is a regulator of immune cells (35). To test whether the observed resistance of 24p3^−/− mice to DSS was indeed conferred by a change in systemic inflammatory response, we performed differential blood cell counts at day 7 post DSS exposure in both groups of mice. There was no significant difference in baseline circulating leukocyte counts in naïve mice as well as in DSS administered mice of both genotypes (Figure 2A). However, upon DSS exposure there was an increase in neutrophil counts in 24p3^+/+ mice but not in 24p3^−/− mice (Figure 2A). In contrast, lymphocyte counts were decreased in DSS administered 24p3^+/+ mice compared to naïve 24p3^+/+ mice (Figure 2A).

Interestingly, exacerbated inflammatory response in DSS-administered 24p3^+/+ mice is not associated with splenomegaly. Based on the organ/body weight ratio, naïve as well as DSS-administered 24p3^+/+ mice displayed no changes in spleen (Figures 2A,B). However, we observed marked splenomegaly in DSS exposed 24p3^−/− mice (Figure 2B). Based on the organ/body weight ratio, the DSS-administered 24p3^−/− mice had ~2.2-fold larger spleens than DSS-administered 24p3^+/+ mice (Figure 2C). We next examined the histological sections of spleens of naïve as well as DSS administered 24p3^+/+ mice and 24p3^−/− mice to gain insight into the observed splenomegaly. While there are no significant changes in the splenic architecture of the 24p3^+/+ mice regardless of treatment, we found that there was significant follicle destruction and infiltration of inflammatory cells in spleens of 24p3^−/− mice treated with DSS (Figure 2D).

24p3 is a bacteriostat therefore it was purported that 24p3 may be required to mount a proper response to prevent overgrowth of commensal microflora in the lumen of the colon in response to intestinal inflammation because of DSS treatment (7). To examine whether resident gut flora were altered in the absence of 24p3 in naïve as well as in DSS administered mice, we performed 16 s rRNA sequencing of fecal samples. We found significantly more bacteria belonging to phylum Proteobacteria in naïve 24p3^−/− mice as compared to naive 24p3^+/+ mice (Figures 3A,B). However, DSS treatment did not cause an increase in these bacteria in 24p3^−/− mice (Figure 3B). Bacteroidetes phylum, was rather decreased upon exposure to DSS in 24p3^+/+ mice but not in DSS-administered 24p3^−/− mice (Figures 3A,B). Firmicutes were increased in DSS-administered 24p3^+/+ mice but not in DSS-administered 24p3^−/− mice (Figures 3A,B). In summary, 24p3^+/+ mice but not 24p3^−/− mice displayed significant changes in composition of bacteria phyla in response to DSS treatment.

Systemic dissemination of bacteria and their components following the loss of epithelial barrier may result in profound changes in cytokine and chemokine responses. To examine this, we measured a variety of cytokines (IL-15, IL-18, bFGF, PDGF-BB, VEGF, TNFα, and LIF) and chemokines (M-CSF, MIP-2, MIG) in sera of DSS administered mice of both genotypes. Among all the components that were analyzed, the amounts of PDGF-BB levels were significantly higher in serum of 24p3^−/−mice relative to DSS administered 24p3^+/+ mice (Figure 4A). In contrast, levels of chemokines and pro-inflammatory cytokines were unaltered despite DSS treatment in 24p3^−/− mice (Figure 4A).

Cytokines locally expressed in the colonic mucosa are differentially expressed following DSS treatment. Therefore, we also evaluated colonic cytokine production in 24p3^+/+ and 24p3^−/− mice challenged with DSS by assessing transcript levels of pdgf-aa, pdgf-bb, tnfα, and vegf (Figure 4B). In agreement with the increased serum PDGF-BB levels, we found that transcript levels of pdgf-bb were also higher in colonic tissue of 24p3^−/− mice administered with DSS compared to DSS administered 24p3^+/+ mice (Figure 4B). Expression levels of pdgf-aa, tnfα, and vegf remain unaltered (Figure 4B). In summary, we found differentially altered expression of PDGF-BB in response to DSS treatment in 24p3^−/− mice.

The hepcidin-ferroportin axis plays an important role in the regulation of iron levels in inflammation (24). Based on structural studies it was found that 24p3 is an iron chelator, but it's in vivo role in systemic iron metabolism is unclear. Interestingly, 24p3 levels were significantly elevated in hepcidin null mice suggesting that it may compensate for hepcidin loss (27). We examined the role of this axis in iron metabolism using DSS treated 24p3^+/+ and 24p3^−/− mice. In agreement with previous studies, we found that hepcidin levels were elevated in DSS treated 24p3^+/+ mice [Figure 5A; (39)]. However, hepcidin levels were lower in DSS treated 24p3^−/− mice (Figure 5A).

Hepcidin leads to hypoferremia and to determine whether differential levels of hepcidin in DSS treated 24p3^+/+ and 24p3^−/− mice contributes to altered serum iron, we measured serum iron indices. We found that total serum iron was lower in DSS administered 24p3^−/− mice (Figure 5B). Transferrin saturation although lower in DSS treated 24p3^−/− mice but did not attain statistically significant levels (Figure 5B). Regression analysis further confirmed an inverse correlation of hepcidin and serum iron (Figure 5C). In addition, we also found that iron levels in the liver and spleen were also lower in DSS administered 24p3^−/− mice (Figure 5D). It is possible that lower systemic iron levels may be one of the pathways for resistance of 24p3^−/− mice to DSS. To test whether exogenous supplementation of iron reverses DSS resistance in 24p3^−/− mice, we injected iron dextran into both 24p3^+/+ and 24p3^−/− mice and subjected them to DSS treatment. As expected naïve 24p3^−/− mice were resistant to DSS. However, exogenous supplementation of iron dextran resulted in an erosion of resistance to DSS in these mice (Figures 5E,F). Interestingly, exogenous administration of iron dextran also conferred enhance sensitivity to DSS administration in 24p3^+/+ mice (Figures 5E,F).

If 24p3 is a direct regulator of pdgf-bb then ectopic addition of 24p3 should neutralize pdgf-bb expression. To test this, we cultured colonic tissues from DSS-administered 24p3^+/+ and 24p3^−/− mice and ectopically added recombinant 24p3 and assessed pdgf-bb transcripts (Figure 6). We detected higher transcripts of pdgf-bb in colonic tissue of DSS administered 24p3^−/− mice (Figure 6A). However, upon addition of recombinant 24p3, we found that pdgf-bb levels were reduced in colonic tissue of DSS administered 24p3^−/− mice (Figure 6A). To additionally confirm these results, we assed pdgf-aa levels and found their levels to remain unchanged (Figure 6A). Thus, these results show that pdgf-bb is negatively regulated by 24p3.

PDGF-BB is found to exhibit protective and mitogenic activities that appear to be essential for the proliferation of cells in the normal gut mucosa as well as for the repair and healing process occurring in the damaged gut mucosa (40). Finally, to directly test the contribution of PDGF-BB in the resistance of 24p3^−/− mice to DSS treatment, we neutralized PDGF-BB by administering anti-PDGF-BB antibody and assessed the survival post DSS administration. In conformity with the results of Figure 4, we detected a higher level of PDGF-BB in DSS administered 24p3^−/− mice (Figure 6B). However, upon injection with an anti-PDGF-BB antibody we observed enhanced sensitivity to DSS in 24p3^−/− mice when compared to control uninjected 24p3^−/− mice (Figure 6C). Thus, these results suggest that increased resistance of 24p3^−/− mice to DSS is in part mediated by PDGF-BB and neutralization of this response abrogates the resistance.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.